Published January 1, 2025 | Version v1

AI-Driven Predictive Handover Management and Energy-Efficient Resource Slicing in Cellular-IoT Empowered C-ITS Environments

Description

The rapid evolution of Cooperative Intelligent Transportation Systems (C-ITS) and 5G-enabled Cellular Internet of Things (Cellular-IoT) ecosystems requires ultra-reliable low-latency communication (URLLC) and massive device connectivity to support safety-critical vehicular services. However, the high mobility of connected and autonomous vehicles (CAVs) introduces severe challenges, including frequent cell handovers, high interruption latencies, and rigid radio resource allocation that fails to adapt to fluctuating traffic densities. Conventional reactive mobility management mechanisms often result in packet loss, connection degradation, and excessive energy consumption in vehicular user equipments (UEs). This paper proposes an integrated AI-driven framework for predictive handover management and energy-efficient resource slicing in Cellular-IoT empowered C-ITS environments. The proposed architecture leverages deep learning and recurrent neural network models to forecast vehicle trajectories and anticipate handover triggers proactively. Simultaneously, a context-aware resource slicing engine dynamically allocates spectrum blocks and optimizes transmission power while integrating 3GPP power-saving mechanisms (such as eDRX and PSM). Simulation-based evaluations demonstrate that the proposed framework significantly reduces handover latency, eliminates unnecessary ping-pong handovers, improves packet delivery ratios, and achieves substantial energy savings compared to conventional reactive and multi-hop baseline schemes.

Files

IJSET_V13_issue6_396.pdf

Files (641.7 kB)

Name Size
md5:4d2a58757ea4a51b71fb07cf4f72983b
641.7 kB Preview Download

Additional details